VERBO: Voice Emotion Recognition dataBase in Portuguese Language
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The recognition of human emotional traits based on Affective Computing is being carried out by computational systems that are able to interpret and react intelligently to the context of the user. Speech Emotion Recognition systems are capable of transforming speech signal data into information related to the feelings of individuals in specific situations. However, the emotional expression of a human being depends mainly on his origins. For this reason, emotional voice databases are peculiar to each language. In this paper, we propose a new emotional database with speech in the Portuguese language of Brazil, called Voice Emotion Recognition dataBase in Portuguese language (VERBO). The database was validated by a panel of expert judges and we achieved an agreement rate of 76% using the content validity index and substantial agreement rate of 65% using Fleiss' Kappa. In addition, an accuracy of 0.76 was achieved and it was possible to observe that the emotions anger and happiness were more easy to recognize showing 0.85 and 0.83 of f1-score, respectively, whereas the disgust and surprise emotions were the most difficult showing 0.67 and 0.68, respectively. In view of this, the main contributions to research made by this study are: (1) The establishment of a new actuated voice database; (2) support provided by voice recognition systems for the analysis of feelings and emotions; and (3) statistical validation of the database using CVI and Fleiss kappa.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it